Intro

Column 1

Music and emotions

Model of emotions, music, and mood

Model of emotions, music, and mood

Causal model

Model of emotions, music, and mood

Model of emotions, music, and mood

Column

Depression

People that scored high on depression listened to higher valence music. This seems contradictory, but research shows that happy music isn’t only for happy people.

Anxiety

Anxious people seem to listen more to contemporary music, but stressed people do not. What can be said about this?

Stress

If you are stressed, what kind of music do you prefer? The stressed people in our research showed some preferences, that might help you unwind as well

Happiness

What kind of music do happy people listen to? Even if you’re happy, you might not know the answer, but Spotify provides us with some interesting insights.

The Data

Column

Data was collected in a survey on the Qualtrics platform Our Questionnaire Structure:

  1. Questions about the demographics

  2. STOMP-R for musical preferences

  3. Participant is asked to provide the link of their personalized Spotify ‘On Repeat’ playlist

  4. Scales about emotional states over the last month: DASS-21 (depression, anxiety and stress) and SHS (happiness)

  5. Participants are asked to indicate how they perceive the emotions reflected by their ‘On Repeat’ playlist, answering the GEMS Emotional Music Scale

  6. Questions about change in listening habits due to quarantine

Column

Age

Country

STOMP-R

Artists

Tracks

Emotions

All

Column

Column

From the theory it follows that there should be some kind of relationship between psychological characteristics and the music people listen to. Although Spotify features are vague and nobody really knows what they mean (including Spotify themselves), they do provide us with interesting statistics on music. To analyse the On Repeat playlists the participants provided us with, we calculated the mean for the Spotify features of every playlist. We then compared these features to the participant’s score on the four psychology questionnaires. What will the songs people listen to on Spotify tell us about their emotional states, and conversely, what will their emotional states tell us about what music they listen to?

To find the patterns, we calculated the correlation between the scores on the questionnaires and the Spotify features. All the significant correlations (below -0.1 and above 0.1) are shown on the left. There seem to be relationships- more happy and anxious people like to listen to more speechy music. Higher depression scores seem to imply lower values of most features, but also liveness- do depressed people dislike concerts or is there something else going on?

Depression

Column

Column

Depression groups

Gender

Column

Research conducted by Stewart et al. (2019) found that young adults use music to deal with depression both through listening to music that differs from their mood (in an attempt to raise spirits), as well as through music that matches their (negative) feelings. Our research showed that those who were depressed tended to avoid low valence songs, with there being a strong correlation between depression and valence.
They also tended to listen to more rock, alternative and punk music.

This can be seen in the graph on the left. People’s score for depression are plotted against their mean valence, with a regression line. As you can see, people scoring higher on depression score higher on valence.

In the second tab, we see the same plot, but with groups for gender.

Anxiety

Column

Column

Column

Our results show that people who are more anxious have been listening more to contemporary music. People with a higher anxiety score reported listening more to funk, reggae, gospel, soul and r&b. Even though previous research has shown that relaxing music can reduce anxiety, we found no relation between anxiety and Spotify features such as energy, loudness and tempo (REF3 & 4).

Interesting finding / Point of discussion: The World Health Organization estimates about 20% of the western pacific region to have an anxiety disorder (REF). However, in our sample, 25% had severe anxiety and 66.7% had extremely severe anxiety. This might be explained by the fact that our survey was conducted during the Corona crisis. Psychologists have speculated about how this crisis could result in an increase in several disorders including anxiety (REF2). One could wonder, though, how quickly people adjust their music listening habits to their state of mind.

Stress

Column

Column

Instrumentalness

Stress and STOMP

Column

Little research was conducted on the stress-reducing effect of music listening on everyday life. Sandstrom & Russo (2010) found that peaceful music (positive valence and low-arousal music) promoted better recovery from an acute stressor. According to Linnemann et al. (2015), relaxation as a reason for music listening was associated with a reduction in stress. Our research has found that people with higher stress levels do not listen to instrumental music, which could be considered as peaceful music. The correlation between the STOMP scores and stress graph shows that people with stress like to listen to classical music, which is considered to be instrumental.

Happiness

Column

Column

Column

First and foremost, despite everything, people still are quite happy! The majority of our sample scored either average or high on happiness. Our hypothesis for happiness was built on the assumption that happiness should be negatively related with the psychopathological states, and depression in particular (REF - also more elaborate hypothesis from that - I think ). Indeed, a negative relationship in our sample was evident (r2=0,21, do we report that?).
The whole ‘negative correlate with intense music’, as opposed to depression? Is it necessary and/or do I talk about this here?
Perhaps unsurprisingly, our data suggest that establishing clear music listening patterns for happy people is quite difficult. One feature that stood out a little was speechiness, namely, it was shown to be higher for people with higher levels of happiness. However, speechiness tends to be generally low in songs, thus any interpretation for this relationship is quite ambiguous (Ref. 2). Furthermore, scores on happiness were found to be positively correlated with energy among males. Women showed no such relationship, and their happiness scores were generally more dispersed. Upbeat, energetic music had been earlier found to be used both for up-regulating (or maintaining) positive emotions as well as down-regulating negative emotions (ref. 3: Cook et al., 2017), thus our findings could be reflecting such patterns. However, we did not make any predictions for gender differences. A speculative explanation could lie in the differences between emotional-regulation strategies in males and females (ref. 4: Thomson et al., 2014), albeit further research is necessary to test that.

Mood

Column

Column

text about mood

Literature

Column

What is known so far?

Music can elicit positive and negative emotions, and can also be a tool for emotion regulation. (Cook et al, 2018)

Studies have found a relationship between various genres of music (rock, metal, hip hop) and behaviours, such as antisocial behaviour, vulnerability to suicide, and drug use. (Scherer et al., 2001)

Music preference is indicative of an underlying emotional disturbance or vulnerability (Scherer et al., 2001)

Through cognitive appraisal, the aesthetic context of music is recognized and emotions are induced (Kawakami, Furukawa, Katahira & Okanoya, 2013).

Another study found out why sad music exists in the first place (Kawakami, Furukawa, Katahira & Okanoya, 2013). They found that when listening to sad music, people’s felt emotions are different from the perceived emotions. One of their explanations for this was that people felt pleasant emotions when listening to sad music due to something called sweet anticipation. People expected to feel sad emotions and due to having successfully predicted this, they experienced pleasant emotions. This comes back to mood adjustment; it implies that people listen to sad music on purpose to feel sad emotions. Two of the same researchers also researched if personality traits relate to people’s consumption of sad music (Kawakami & Katahira, 2015). They found that several sub-components of the personality trait empathy were positively related to the consumption of sad music.

column

Kawakami, A., Furukawa, K., Katahira, K., & Okanoya, K. (2013). Sad music induces pleasant emotion. Frontiers in Psychology, 4. doi: 10.3389/fpsyg.2013.00311

Kawakami, A., & Katahira, K. (2015). Influence of trait empathy on the emotion evoked by sad music and on the preference for it. Frontiers in Psychology, 6. doi: 10.3389/fpsyg.2015.01541